BUG: support "fill_value" for ".unstack()" called with list of levels

closes #13971

Author: toobaz

doc/source/whatsnew/v0.21.0.txt

@@ -1008,6 +1008,7 @@ Reshaping
 - Bug in :func:`concat` where order of result index was unpredictable if it contained non-comparable elements (:issue:`17344`)
 - Fixes regression when sorting by multiple columns on a ``datetime64`` dtype ``Series`` with ``NaT`` values (:issue:`16836`)
 - Bug in :func:`pivot_table` where the result's columns did not preserve the categorical dtype of ``columns`` when ``dropna`` was ``False`` (:issue:`17842`)
+- Bug in :func:`unstack` which, when called on a list of levels, would discard the ``fillna`` argument (:issue:`13971`)
 
 Numeric
 ^^^^^^^

pandas/core/reshape/reshape.py

@@ -292,7 +292,7 @@ def get_new_index(self):
                           names=self.new_index_names, verify_integrity=False)
 
 
-def _unstack_multiple(data, clocs):
+def _unstack_multiple(data, clocs, fill_value=None):
     if len(clocs) == 0:
         return data
 
@@ -330,7 +330,7 @@ def _unstack_multiple(data, clocs):
     if isinstance(data, Series):
         dummy = data.copy()
         dummy.index = dummy_index
-        unstacked = dummy.unstack('__placeholder__')
+        unstacked = dummy.unstack('__placeholder__', fill_value=fill_value)
         new_levels = clevels
         new_names = cnames
         new_labels = recons_labels
@@ -347,7 +347,7 @@ def _unstack_multiple(data, clocs):
         dummy = data.copy()
         dummy.index = dummy_index
 
-        unstacked = dummy.unstack('__placeholder__')
+        unstacked = dummy.unstack('__placeholder__', fill_value=fill_value)
         if isinstance(unstacked, Series):
             unstcols = unstacked.index
         else:
@@ -460,7 +460,7 @@ def unstack(obj, level, fill_value=None):
         if len(level) != 1:
             # _unstack_multiple only handles MultiIndexes,
             # and isn't needed for a single level
-            return _unstack_multiple(obj, level)
+            return _unstack_multiple(obj, level, fill_value=fill_value)
         else:
             level = level[0]
 

pandas/tests/frame/test_reshape.py

@@ -154,6 +154,29 @@ def test_unstack_fill(self):
                              index=['x', 'y', 'z'], dtype=np.float)
         assert_frame_equal(result, expected)
 
+        # GH #13971: fill_value when unstacking multiple levels:
+        f = DataFrame({'x': ['a', 'a', 'b'],
+                       'y': ['j', 'k', 'j'],
+                       'z': [0, 1, 2],
+                       'w': [0, 1, 2]}).set_index(['x', 'y', 'z'])
+        unstacked = f.unstack(['x', 'y'], fill_value=0)
+        key = ('w', 'b', 'j')
+        expected = unstacked[key]
+        result = pd.Series([0, 0, 2], index=unstacked.index, name=key)
+        assert_series_equal(result, expected)
+        stacked = unstacked.stack(['x', 'y'])
+        stacked.index = stacked.index.reorder_levels(f.index.names)
+        # Workaround for GH #17886:
+        stacked = stacked.astype(int)
+        result = stacked.loc[f.index]
+        assert_frame_equal(result, f)
+
+        # From a series
+        s = f['w']
+        result = s.unstack(['x', 'y'], fill_value=0)
+        expected = unstacked['w']
+        assert_frame_equal(result, expected)
+
     def test_unstack_fill_frame(self):
 
         # From a dataframe